Set Analysis Syntax

[cbaqir]

My data set shows transactional records that may result in multiple rows for the same workflow step and this it repeats the workflow duration.

REQUEST_ID , WORKFLOW_STEP_NAME , WORKFLOW_DURATION, TRX_DATE

01, StepA, 66, 12/31/2014

02, StepA, 4, 11/30/2014

03, StepA, 8, 7/31/2014

04, StepA, 2, 9/25/2014

05, StepA, .123, 8/5/2014

06, StepA, 1, 11/12/2014

];

I need to show the Avg Workflow Duration but don't want to include the duplicate rows:

i.e.

REQUEST_ID , WORKFLOW_STEP_NAME , WORKFLOW_DURATION, TRX_DATE

01, StepA, 66, 10/31/2014

01, StepA, 66, 11/30/2014

01, StepA, 66, 12/31/2014

That being said, it's possible that a request repeats a workflow step - for example, we would see two rows with the same workflow step name but with different values for workflow duration.

REQUEST_ID , WORKFLOW_STEP_NAME , WORKFLOW_DURATION, TRX_DATE

01, StepA, 5, 8/30/2014

01, StepA, 66, 10/31/2014

01, StepA, 66, 11/30/2014

01, StepA, 66, 12/31/2014

];

Neither of these is exactly right when compared with the raw data.

QV is showing an average of ~24 days but if you exclude two of the entries for Req 01 for 66 days, it should be ~12 days.

Any suggestions? TIA!

[cbaqir]

QV is showing an average of ~24 days but if you exclude two of the entries for Req 01 for 66 days, it should be ~12 days.

[cbaqir]

That's great, Manoj! I've never seen anything quite like what you did and I think it will work for me. Thanks!